Abstract
•The secure construction of 6TiSCH network is implemented by performing authentication of nodes based on security credentials. The nodes are formed into corona based DODAG structure to reduce overhead.•The dynamic selection of parent node is performed by implementing RCM optimizer based on distance, link quality, expected transmission count, trust and buffer rate. This improves the transmission reliability in the network.•The critical traffic aware optimized scheduling is carried out by using SGD and XGBoost keeping in mind the criticality of heterogeneous tasks.The risk aware secure offloading is executed in which the multiple tasks are offloaded to multiple edge servers in order to achieve balance of load, energy efficiency.
The evolution of 6TiSCH networks is greatly increasing in the field of IIoT which supports reliable communication in IIoT environments. However, there is a lot of research gaps in the field of IIoT using 6TiSCH networks in terms of high latency, more energy consumption which degrades the overall performance of the IIoT environment. To address the above limitation in existing IIoT 6TiSCH approaches, an Edge-assisted 6TiSCH network for IIoT is proposed to overcome the above challenges such as Low latency and more energy consumption by efficient task scheduling and edge offloading considering risk in IIoT environment. The proposed work consists of three layers namely the 6TiSCH layer, Edge layer, and Cloud layer. In the 6TiSCH layer, initially, devices are registered by their credentials and authenticated by TA using Enhanced Advanced Encryption Standard (EAES) which reduces the unwanted energy consumption and latency by dropping the malicious devices. After authentication, only the verified devices are allowed for network construction using CORONA-based DODAG construction for selecting optimal parent which reduces the energy consumption. The optimal parent is selected by using Red Colobuses Monkey (RCM) optimization algorithm. Then two-level task scheduling is done namely Slot frame length optimization using Stochastic Gradient Descent (SGD) algorithm and adaptive partitioning using XG boost algorithm. In the Edge layer, effective offloading is done by using the Soft Actor-Critic (SAC) algorithm for reducing the energy consumption during offloading. Finally, all the processed data are sent to the cloud layer for access. The proposed work is experimented with using the Cooja simulation tool with Contiki 3 OS. The performance of the proposed work is compared to the existing works in terms of Latency (42 ms-20.4 ms) low, energy consumption (28.8 J-12 J) low in 6TiSCH layer, Latency (40 ms-20.4 ms) low, energy consumption (24.8 J-13.6 J) low in Edge layer, and overall comparison achieves throughput (508.8Kbps-322Kbps) high, packet delay ratio (8 %-4.6 %) high, end-to-end delay (6 s-9 s) low and efficiency (14.6 %-7.8 %) high. The experimental results show that our proposed work performs well than existing works.